Your files stay on your device

Prompt Diff: Token and Cost Delta Between Two Prompts

Before/after prompts with exact token deltas, a line diff, and the cost change per model.
exact token counts both sides · line diff · cost delta per model · worker-only, nothing leaves the tabClearEncodingCost delta models (up to 4)gpt-5.6-solgpt-5.6-terragpt-5.6-lunagpt-5.6-cybergpt-5.5gpt-5.5-progpt-5.4gpt-5.4-minigpt-5.4-nanogpt-5.4-progpt-5.3-codexgpt-5.2gpt-5.2-progpt-5.1gpt-5gpt-5-minigpt-5-nanogpt-5-progpt-4.1gpt-4.1-minigpt-4.1-nanogpt-4ogpt-4o-minio3o3-minio3-proo4-minio1chat-latestgpt-4-turbogpt-3.5-turbo

1 selected. Deltas are exact nano-USD at the registry rates, zero output tokens.
Compare

- Both sides counted exactly with the selected encoding. No chars-÷-4 estimates anywhere on this page.
- The line diff reuses the audited text-diff engine; tokenization runs in the AI family worker so vocabulary tables never enter the page.
- Cost deltas use the dated registry rates at zero output tokens: they isolate what the prompt change alone does to the input bill.
- Each side is capped at 60,000 characters; prompts do not leave this tab.
Nothing in yet

Paste, drop, or type to begin. Everything stays on this device.

Next door in AI

- Tokenizer playground (/ai/tokenizer)
- API cost estimator (/ai/cost-estimator)
All AI tools (/ai)

Canonical HTML: https://nutter.tools/ai/prompt-diff
Markdown version: https://nutter.tools/ai/prompt-diff/index.md
Plain-text version: https://nutter.tools/ai/prompt-diff/index.txt
Agent index: https://nutter.tools/llms.txt